I would like to thank the faculty and staff of the Kellogg Radiation Laboratory for the stimulating and friendly environment they have provided. The 700 KeV Van de Graaff electrostatic generator at the Kellogg Radiation Laboratory was used to produce a hydrogen ion beam whose energy was analyzed by a 90° bending magnet system (Davids, 1968). These were mounted on separate rotating arms spaced 8.5 em from the axis of the chamber.
The solid angle of the detectors was approximately 10 -3 steradians, defined by a 3 mm round hole in the tantalum disk in front of each detector. In addition to definitely confirming the existence of the 470 KeV resonance, these results also confirm the existence of the previously reported anomaly at about 680 KeV. The raw data for the excitation functions were the count rates in the detectors set to elab = 85 0 or 90. • Results were corrected to 0 for dcrcm/dalab (Appendix, Part A) and corrected to e --em 90° using i .
The raw count rates for beryllium scattering were multiplied by E, where E is the average 2 alpha energy in the beryllium target layer. The function S to destroy Be using the sum of the sections (p,d)9 and (p,a) from Figures 9 and 10 is presented in the figure.
RANDOM ERRORS
The energies of the incoming alpha particles used in Rutherford scattering were uncertain by ±1%. The thicknesses of the 50 KeV thick targets were measured by recording the shifts in energy of the elastically scattered alpha peak caused by placing the target in front of the detector. The total uncertainty in the conversion of count rates to absolute cross sections using Rutherford scattering measurements is estimated by incoherently summing the errors from parts B and D to G.
The unnormalized counting rate of reaction products·at the peak of the 330 KeV resonance was determined to ±2%, thus the absolute crossover. All deuteron count rates are normalized with respect to alpha count rates in the same detector. There is an estimated uncertainty of ±3% in the normalization of the thick target count rates with respect to thin target rates at the same energies (Figures 9 and 10).
This gave the thickness of 9Be on the fixed targets in the beam direction as 10± 1.5 ~g/cm 2 • This uncertainty in the thickness, together with the uncertainties introduced by the approximations in the appendix, results in a varying uncertainty in the correction factor applied to the low energy count rates. The effects discussed in parts L and M are included in the error bars in Figures 9–10 and those from L, M, and N in Figure 12 .
THEORETICAL ANALYSIS
The above approximations were made to simplify the analytical form of the integrals in Tdirect. The three channel interaction radii and the reduced pole width amplitudes and energies were parameters that were modified to match the experimental cross sections. Because of the many parameters involved in the interaction of multiple broad resonances, it is not possible to determine a unique fit to the experimental cross sections and angular distributions.
The 330 KeV resonance, the most prominent feature of the low-energy cross sections, was fitted with a 1, level in the 10 B junction core. The size of the cross sections at the resonance and at a laboratory energy of 50 to 100 KeV were the determining factors for making the fit. An upper limit of Q ~ 10 (see Appendix G for a definition of Q} for. p,d) the response was estimated by calculating the amplitude of.
The change with energy of the direct reaction cross section is almost the same as the observed total cross section below 300 KeV, so direct reactions cannot cause low-energy enhancements in d. Because of the large asymmetry in the cross-sections above 400 KeV and the definite resonance shape observed in low-angle production near 470 KeV (laboratory), the excited state at this energy is almost certainly of positive parity. The parameters used for this level did not correspond to the behavior of a , asymmetry a or d .
The beam spot was observed (by discoloration of the target) to lie 1.4 mm to one side of the axis of the target chamber. The raw angular distribution data were corrected for the varying distances between the detectors from the beam spot. This reduction was due to the charge state of the proton beam being changed in the target and scattering of a fraction of the incident beam out of the solid angle of the Faraday cup.
During the run-in period, a significant thickness of carbon would build up on the surface of the target. Because of the energy lost to the target by the incoming protons, the measured count rate was less than would be measured if the beam. The inverse of the quantity that multiplies da(Ei) in equation F.2 is multiplied by the observed count rate to give a rate proportional to the true cross section at E = Ei.
The choice of this form for Vext. causes all the solid angle arguments of the spherical harmonics in G.7 to G.9 to be identical. Thus, the angular integration in the integral for Tdirect reduces to an integral of a product of three spherical harmonics of the same argument:. The first two columns give the energy (in MeV) and JTI values of the levels.
The excitation functions for 9Be(p,a) and (p,d) • The table data are: the proton laboratory energy in KeV, the (p,a) total cross-section in millibarns, the (p,d) cross-section and S in MeV stable as defined.
Ep KeV
Geometry of the target showing the relationship of the beam spot to the scattering chamber axis. Ratio of the jet flux measured at the Faraday cup to the flux hitting the target as a function of the proton laboratory energy. The ratio of the actual cross-section to the measured cross-section as a function of the proton laboratory energy for a target thickness of 22 ~g/cm2• See page 27.
The ratio between real cross-section and measured cross-section as a function of proton laboratory energy for a target thickness of 10 ~g/cm 2 • See page 27.
